[PATCH] perf: Force USER_DS when recording user stack data.

[PATCH] perf: Force USER_DS when recording user stack data.

[Yabin Cui]

Perf can record user stack data in response to a synchronous request, such
as a tracepoint firing. If this happens under set_fs(KERNEL_DS), then we
end up reading user stack data using __copy_from_user_inatomic() under
set_fs(KERNEL_DS). I think this conflicts with the intention of using
set_fs(KERNEL_DS). And it is explicitly forbidden by hardware on ARM64
when both CONFIG_ARM64_UAO and CONFIG_ARM64_PAN are used.

So fix this by forcing USER_DS when recording user stack data.

Signed-off-by: Yabin Cui <yabinc@google.com>
---
 kernel/events/core.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/kernel/events/core.c b/kernel/events/core.c
index 2a62b96600ad..9bc047421e75 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -5948,6 +5948,7 @@ perf_output_sample_ustack(struct perf_output_handle *handle, u64 dump_size,
 		unsigned long sp;
 		unsigned int rem;
 		u64 dyn_size;
+		mm_segment_t fs;
 
 		/*
 		 * We dump:
@@ -5965,7 +5966,10 @@ perf_output_sample_ustack(struct perf_output_handle *handle, u64 dump_size,
 
 		/* Data. */
 		sp = perf_user_stack_pointer(regs);
+		fs = get_fs();
+		set_fs(USER_DS);
 		rem = __output_copy_user(handle, (void *) sp, dump_size);
+		set_fs(fs);
 		dyn_size = dump_size - rem;
 
 		perf_output_skip(handle, rem);
-- 
2.19.0.rc0.228.g281dcd1b4d0-goog

Re: [PATCH] perf: Force USER_DS when recording user stack data.

[Peter Zijlstra]

On Thu, Aug 23, 2018 at 03:59:35PM -0700, Yabin Cui wrote:
> Perf can record user stack data in response to a synchronous request, such
> as a tracepoint firing. If this happens under set_fs(KERNEL_DS), then we
> end up reading user stack data using __copy_from_user_inatomic() under
> set_fs(KERNEL_DS). I think this conflicts with the intention of using
> set_fs(KERNEL_DS). And it is explicitly forbidden by hardware on ARM64
> when both CONFIG_ARM64_UAO and CONFIG_ARM64_PAN are used.
> 
> So fix this by forcing USER_DS when recording user stack data.
> 
> Signed-off-by: Yabin Cui <yabinc@google.com>

Ingo, I think this wants a stable tag too; seems to be a corrolary of:

  88b0193d9418 ("perf/callchain: Force USER_DS when invoking perf_callchain_user()")

Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>

> ---
>  kernel/events/core.c | 4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/kernel/events/core.c b/kernel/events/core.c
> index 2a62b96600ad..9bc047421e75 100644
> --- a/kernel/events/core.c
> +++ b/kernel/events/core.c
> @@ -5948,6 +5948,7 @@ perf_output_sample_ustack(struct perf_output_handle *handle, u64 dump_size,
>  		unsigned long sp;
>  		unsigned int rem;
>  		u64 dyn_size;
> +		mm_segment_t fs;
>  
>  		/*
>  		 * We dump:
> @@ -5965,7 +5966,10 @@ perf_output_sample_ustack(struct perf_output_handle *handle, u64 dump_size,
>  
>  		/* Data. */
>  		sp = perf_user_stack_pointer(regs);
> +		fs = get_fs();
> +		set_fs(USER_DS);
>  		rem = __output_copy_user(handle, (void *) sp, dump_size);
> +		set_fs(fs);
>  		dyn_size = dump_size - rem;
>  
>  		perf_output_skip(handle, rem);
> -- 
> 2.19.0.rc0.228.g281dcd1b4d0-goog
> 

[tip:perf/urgent] perf/core: Force USER_DS when recording user stack data

[tip-bot for Yabin Cui]

Commit-ID:  02e184476eff848273826c1d6617bb37e5bcc7ad
Gitweb:     https://git.kernel.org/tip/02e184476eff848273826c1d6617bb37e5bcc7ad
Author:     Yabin Cui <yabinc@google.com>
AuthorDate: Thu, 23 Aug 2018 15:59:35 -0700
Committer:  Ingo Molnar <mingo@kernel.org>
CommitDate: Mon, 10 Sep 2018 14:01:46 +0200

perf/core: Force USER_DS when recording user stack data

Perf can record user stack data in response to a synchronous request, such
as a tracepoint firing. If this happens under set_fs(KERNEL_DS), then we
end up reading user stack data using __copy_from_user_inatomic() under
set_fs(KERNEL_DS). I think this conflicts with the intention of using
set_fs(KERNEL_DS). And it is explicitly forbidden by hardware on ARM64
when both CONFIG_ARM64_UAO and CONFIG_ARM64_PAN are used.

So fix this by forcing USER_DS when recording user stack data.

Signed-off-by: Yabin Cui <yabinc@google.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: <stable@vger.kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: 88b0193d9418 ("perf/callchain: Force USER_DS when invoking perf_callchain_user()")
Link: http://lkml.kernel.org/r/20180823225935.27035-1-yabinc@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
 kernel/events/core.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/kernel/events/core.c b/kernel/events/core.c
index abaed4f8bb7f..c80549bf82c6 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -5943,6 +5943,7 @@ perf_output_sample_ustack(struct perf_output_handle *handle, u64 dump_size,
 		unsigned long sp;
 		unsigned int rem;
 		u64 dyn_size;
+		mm_segment_t fs;
 
 		/*
 		 * We dump:
@@ -5960,7 +5961,10 @@ perf_output_sample_ustack(struct perf_output_handle *handle, u64 dump_size,
 
 		/* Data. */
 		sp = perf_user_stack_pointer(regs);
+		fs = get_fs();
+		set_fs(USER_DS);
 		rem = __output_copy_user(handle, (void *) sp, dump_size);
+		set_fs(fs);
 		dyn_size = dump_size - rem;
 
 		perf_output_skip(handle, rem);